The impacts of COVID-19 on transportation sector have received a substantial research attention, however, less is known about localized COVID-19 responses that provided safe space for mobility and other daily activities. We applied logistic regression and text mining approaches on the Shifting Streets COVID-19 Mobility Dataset to explore the long-term outcomes of the localized responses. We explored the purpose, affected space, function, and implementation approach. We found that responses instituted for economic recovery and public health are less likely to be long-term, while responses meant to improve safety or bicycle/pedestrian mobility are more likely to be long-term. Further, operational or regulatory responses are less likely to be long-term. Additionally, responses affecting curb space are more likely to be long-term than those affecting other right-of-way areas. Text-mining of responses’ narratives revealed key patterns for both short-term and long-term outcomes. Study findings showcase the possible design and operations changes during post-COVID-19 era.